551–560 of 1001 posts
So how long before AMD and Intel start packing 16GB of RAM on their chips? The M1 has what, 1.5x the bandwidth of the other top chips?
I know very little, so perhaps someone could enlighten me. But I am curious how Apple Silicon will be for machine learning. When Apple releases a MacBook Pro with 64GB of unified memory (assuming they will) — won’t that be amazing for machine learning? I am under the impression that GPU memory is a huge factor in performance. Also, is there any way that the neural engine can accelerate training — or is it just for ex…
I wouldn't expect it to be particularly competitive in training large models. It's an integrated GPU with 8 cores, and the "neural engine" has an additional 16 cores. The kinds of discrete GPUs (mostly Nvidia) that people use for deep learning have more like 5000+ cores. I think Apple is aiming more at either training small models, or running pre-trained models. For example Photoshop is starting to integrate neural f…
Earlier quoted context omitted.
> Apple's walled garden Apparently over the years the definition of "Walled Garden" has drifted a lot. The iPhone has a "Walled Garden", unless you jailbreak, it's very difficult to run anything outside the App Store. My Mac? Almost nothing I run is from the App Store. Nothing needs to be from the App Store. Most of what I run doesn't even go through Gatekeeper and it certainly never touches software I build or compi…
The parent was complaining about being forced to use OS X. So, the "walled garden" is intended for the underlying hardware.
I don't think that is what he was saying at all.
Regardless.
There is nothing preventing Linux (or Windows) from booting on Mac M1 hardware. People almost certainly will have Linux running on Mac hardware before too long. It's just a slog getting it working well.
Not a walled garden in any traditional sense, just difficult to implement.
This is fascinating: > Retain and release are tiny actions that almost all software, on all Apple platforms, does all the time. ….. The Apple Silicon system architecture is designed to make these operations as fast as possible. It’s not so much that Intel’s x86 architecture is a bad fit for Apple’s software frameworks, as that Apple Silicon is designed to be a bespoke fit for it …. retaining and releasing NSObjects i…
This quote doesn’t really cover why M1 macs are more efficient with less ram than intel macs? You’ve got a memory budget, it’s likely broadly the same on both platforms, the speed at which your retains/releases happen isn’t going to be the issue. it’s not like intel macs use GC where m1 uses RC. (It explains why iOS does better with less ram than android, but the quote is specifically claiming this as a reason for 8G…
"iOS software uses reference counting for memory management, running on silicon optimized to make reference counting as efficient as possible; Android software uses garbage collection for memory management, a technique that requires more RAM to achieve equivalent performance." This statement is nonsense. Reference counting is typically used in garbage collection. https://en.m.wikipedia.org/wiki/Reference_counting It…
"iOS software uses reference counting for memory management, running on silicon optimized to make reference counting as efficient as possible; Android software uses garbage collection for memory management, a technique that requires more RAM to achieve equivalent performance." This statement is nonsense. Reference counting is typically used in garbage collection. https://en.m.wikipedia.org/wiki/Reference_counting It…
No, they are not equivalent.
It is true that reference counting is a form of GC. However, Java's GC is not based primarily on reference counting. It is much more complex and, generally, does indeed use much more memory.
Swift's reference counting is not much different than C++ shared pointers, except that it is all baked into the language. It is generally true that iOS devices require less memory to achieve the same things as Android devices.
Oh no crap Sherlock! Let me save this quote whenever someone wants to tell me all about how GC is much superior than reference counting.
Yes tell me how a stop-scan-mark-sweep periodic process is more efficient than just keeping track of what you do.
This is fascinating: > Retain and release are tiny actions that almost all software, on all Apple platforms, does all the time. ….. The Apple Silicon system architecture is designed to make these operations as fast as possible. It’s not so much that Intel’s x86 architecture is a bad fit for Apple’s software frameworks, as that Apple Silicon is designed to be a bespoke fit for it …. retaining and releasing NSObjects i…
This quote doesn’t really cover why M1 macs are more efficient with less ram than intel macs? You’ve got a memory budget, it’s likely broadly the same on both platforms, the speed at which your retains/releases happen isn’t going to be the issue. it’s not like intel macs use GC where m1 uses RC. (It explains why iOS does better with less ram than android, but the quote is specifically claiming this as a reason for 8G…
Perhaps they never really needed to fit 32GB into their intel macs either. Some days after the glowing reviews; and strange comments about magic memory utilization; we now see comments concerned about SSD wear due to swap file usage.
If the applications and data structures are more compact in memory on the arm processors; it should be easy to test; you just need an intel mac; and an M1 mac running the same app on the same document and look at how much memory it uses.
Earlier quoted context omitted.
Setting aside that you are way off thread here... > how do I assign >16GB of memory to it if I only have 8-16GB of RAM? This is Apple's slowest/ lowest performance M series CPU. Complaining that the CPU they built for the MacBook Air and the lowest end MacBook Pro doesn't have 32GB of RAM misses the entire picture. This is Apple's first and lowest end M series chip, and it's blowing away Intel chips with discrete GPU…
> In fact since Apple is migrating the entire line-up, it's likely the next generation of CPUs will support discrete RAM I've been wondering about how much of the general purpose performance boost of M1 is due to having the RAM in the same package. That has to have benefits in power and latency. So if a future Mx chip supports discrete RAM, it may not seem quite as magical anymore. Then again, Apple's volume and marg…
They know the exact latencies and can distribute the memory between CPU and GPU as they please.
A loss in upgradeability is a huge gain in speed and reliability.
My bet is that the next M processor will just have more of everything. More cores and more built-in memory. Maybe the one for the (i)Mac Pro will have upgradeable memory on top of the built-in ones. All of the laptops will only have the on-package memory.
You know the thing I worry about next: how are apps going to inevitably bloat in inefficiency and claw back the improvements in CPU?
And then Apple will make us buy newer and better and faster models. Now, we can't even upgrade RAM